The Unicore Grid and Its Options for Performance Analysis

Haubold, Sven; Mix, Hartmut; Nagel, Wolfgang E.; Romberg, Mathilde

doi:10.1007/978-1-4615-0361-3_17

Cited by 5 publications

(4 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The extension includes a group concept allowing the distinction between different metahosts during an analysis session. VAMPIR has also been successfully integrated with the UNICORE grid middleware [10], allowing the user to create a task with appropriate instrumentation and to retrieve the generated trace files for local visualization after program termination. Another grid-enabled performance monitoring and analysis system is SCALEA-G [16].…”

Section: Related Workmentioning

confidence: 99%

Automatic Trace-Based Performance Analysis of Metacomputing Applications

Becker

Wolf

Frings

et al. 2007

2007 IEEE International Parallel and Distributed Processing Symposium

View full text Add to dashboard Cite

The processing power and memory capacity of independent and heterogeneous parallel machines can be combined to form a single parallel system that is more powerful than any of its constituents. However, achieving satisfactory application performance on such a metacomputer is hard because the high latency of inter-machine communication as well as differences in hardware of constituent machines may introduce various types of wait states. In our earlier work, we have demonstrated that automatic pattern search in event traces can identify the sources of wait states in parallel applications running on a single computer. In this article, we describe how this approach can be extended to metacomputing environments with special emphasis on performance problems related to inter-machine communication. In addition, we demonstrate the benefits of our solution using a real-world multi-physics application.

show abstract

Section: Related Workmentioning

confidence: 99%

Automatic Trace-Based Performance Analysis of Metacomputing Applications

Becker

Wolf

Frings

et al. 2007

2007 IEEE International Parallel and Distributed Processing Symposium

View full text Add to dashboard Cite

show abstract

“…The co-scheduling of jobs on different clusters in the VIOLA grid is managed by the grid middleware UNI-CORE [7] which has been enhanced by adding a metascheduler for the simultaneous allocation of compute and network resources. Bierbaum et al [2] describe this UNICORE-based infrastructure supporting the co-allocation of metacomputing resources in more detail, with special emphasis on the intricate task of coordinating network allocation with application startup.…”

Section: Middlewarementioning

confidence: 99%

Performance Evaluation and Optimization of Parallel Grid Computing Applications

Becker

Frings

Wolf

2008

16th Euromicro Conference on Parallel, Distributed and Network-Based Processing (PDP 2008)

View full text Add to dashboard Cite

The combination of independent and potentially heterogeneous parallel machines creates a powerful metacomputer. Such a metacomputer can be used to run a single parallel application if a single machine does not provide enough CPUs. However, achieving satisfactory application performance on such a metacomputer is difficult since instances of grid-related as well as non grid-related performance properties may introduce various wait states during communication and synchronization. In our earlier work, we have introduced an extension to the SCALASCA tool set for recording event traces of metacomputing applications and searching them automatically for patterns of inefficient behavior related to wide-area communication. Here, we show how this extension in combination with statistical analyses and time-line visualization provided by VAMPIR can be applied to evaluate and optimize the performance of a multi-physics production code running on a heterogeneous and geographically dispersed metacomputer.

show abstract

“…Many Grid brokering systems exist, including Condor [3], gLite [4] and Maui [11] of which Condor is one of the most popular. Although Condor does not directly support the OGF standards for resource advertising and job submission it is possible to submit tasks from Condor into a number of different Grid frameworks including, Globus [14], Nor-duGrid [15], UNICORE [2], PBS [16] and LSF [17]. This has been achieved by providing grid specific adapters that sit between the condor system and the remote grid system.…”

Section: Related Workmentioning

confidence: 99%

“…to advertise, submit, select, monitor and transfer files. Vendors exist such as (UNICORE [2], Globus [14], Condor [3], gLite [4]) each providing a customised solution to these problems. However, in general these solutions present a tie-in to their particular interfaces.…”

Section: Introductionmentioning

confidence: 99%

Adding Standards Based Job Submission to a Commodity Grid Broker

Colling

McGough

Smith

et al. 2008

2008 IEEE Fourth International Conference on eScience

View full text Add to dashboard Cite

The Condor matchmaker provides a powerful mechanism for optimally matching between user task and resource provider requirements, making the Condor system a good choice for use as a meta-scheduler within the Grid. Integrating Condor within a wider Grid context is possible but only through modification to the Condor source code as each new mechanism for connection to remote resources is defined. In this paper we describe how the emerging standards for job submission and resource description can be integrated into the Condor system, thereby allowing arbitrary remote Grid resources which support these standards to be brokered using Condor.

show abstract

The Unicore Grid and Its Options for Performance Analysis

Cited by 5 publications

References 3 publications

Automatic Trace-Based Performance Analysis of Metacomputing Applications

Automatic Trace-Based Performance Analysis of Metacomputing Applications

Performance Evaluation and Optimization of Parallel Grid Computing Applications

Adding Standards Based Job Submission to a Commodity Grid Broker

Contact Info

Product

Resources

About